Arm updates

[Nixuelle]

I plotted your data (and the corresponding std. devs.) in OriginLAB and fitted the data to the best matching functions so we could project how long it will take your arms to heal.  In these plots, I've listed the number of months passed as an arbitrary index for the abscissa of the plot, starting at zero from February (where you first started collecting data).  I then plotted the pain index you report (in arbitrary units mind you, since this is all subjective) for both the left and right hands as the ordinate.

Surprisingly, the data fits fairly well to a linear line fitting (I was expecting it to more closely follow an exponential decay).  Correlation coefficients in both cases are above 0.90, better than is normally obtained for lots of biological data in the science world!  Interestingly, from what you've posted and the projections, both of your arms will be back to normal in a little over 5 months from your last reported data points, assuming the rate of improvement remains constant.  This means, if this crude model works, your hands should be back to 100% in early September, Vir!

[Virmir]

Ha ha, that's awesome!  It will be interesting to see if upcoming months fit this model.

I think it's a little funny your plan of attack is to suddenly return if there's no improvement? I kinda get what you're saying, but risking serious injury still seems like a less than ideal plan any way I look at it. Though I'd be lying if I didn't anticipate you coming back so much... |D

I just hope your arm continues to improve.

That's if I'm convinced resting the arms will not make them better, considering we are on month number 11 now.  I'd rather deal with pain and continue on with my life than remain at a standstill with somewhat less pain.  I definitely won't do a full scale comeback immediately but rather a build up over a few months.  These next two months would have to plateau or get worse for me to consider this, though.

No matter the direction I choose, I plan on alternating drawing with one arm one day and the other the next day so in the end I'll only be drawing at most 50% my previous workload with either arm even if I go full tilt.

[Virmir]

Updated monthly arm annoyance factors!  Note that last month I calculated the left hand a day too early and that last day actually bumped it up a little.  Again, keep in mind the right arm is nearly constantly at rest in the left arm in use. (Standard deviations in parentheses.)

Code: [Select]

L R
February 3.1 (1.4) 4.1 (1.8)
March 2.4 (0.9) 3.2 (1.3)
April 2.3 (0.9) 2.9 (1.0)
May 2.4 (0.9) 2.3 (0.9)
Right arm-- noticeable improvement this month.  Still have some issues trying to step up some of the the exercises I do, but at least it's feeling better at rest.  For the month of June I am going to add one full hour weekly drawing session, bringing the total number of weekly right handed hours to 2.

Left arm-- looking back, it looks like there hasn't been much improvement over the past couple of months (although the numbers from the beginning of this month were a little high due to an incident and it has since improved).  I'm pretty confident these short drawing sessions aren't irritating it, but going to take it easy for one more month and then decide.  No new drawing sessions for the left this month (total weekly hours remains at 2).

This month I'm going to start scheduling different right and left handed days (instead of doing 30 minutes with the right followed by 30 minutes with the left like I've been doing).  Eventually I will add more hours to these days so that I will always have at least one day of rest in between drawing sessions for either hand.

Sunday - L 1h
Monday - R 30m
Tuesday - L 30m
Wednesday - (odd day)
Thursday - R 30m
Friday - L 30m
Saturday - R 1h

Total weekly drawing hours for June - 4

[Nixuelle]

Here is a new analysis for both arms, given the added data.  

As can be seen from the above plots, the earlier model for the right arm remained quite accurate with the addition of the new data point.  The predicted value for recovery for the right arm remains approximately the same as predicted in the plot I made last month.  The left arm, however, was more complicated, due to the fact that the new data point did not follow the earlier model.  This resulted in the linear model for the left arm being thrown off significantly, and a bad line fit (R squared value for the linear fit dropped to 0.11; very poor).  

Considering the above, I tried two alternative models for the left arm, including both an exponential and logarithmic decay. Though both curves fit the data fairly well, the exponential model is quite amusing in that it predicts Virmir's left hand will never recover (obviously, this is probably not the case, ha ha!).  Looking at the logarithmic model, however, we see the R squared value indicates that it fits better than the exponential one.  Surprisingly also, the new logarithmic model predicts a recovery that is closely in line with what the earlier linear model from last month predicted.  Given these observations, at present it appears the logarithmic model best predicts the behavior of the left arm.

More observations in the coming months!

[DigitFiredrake]

Let's just hope the exponential one is wrong!!

[Tsunami]

Your pain rating goes from 0 to 10 but nothing more than a 4 typically, so a 4 stops you?   Though I'm a bad example as I've experienced 8s, 9s and 10s before. 

Fall 2003 - One foot of intestine died.  Was an automatic 10 for at least an hour before the doctor got to see me.  For about 3 hours before it, the pain ramped up from 6 to 10.  10 is your entire existence is nothing but physical pain.

June 2014 - pinched a nerve in my lower spine due to herniated disc.  For 9 months I didn't walk much as it would bring incredibely agonizing pain.  It would ramp up quickly after 100 to 200 yards I would be an 8 on the pain scale (that's the point where I have to sit down or throw up wherever I am).  When they MRIed it, that was hell.  Thanks to the way they had to position the legs it turned the pain into a 9.  The pain would come with waves like an ocean.  As the waves came in my awareness of the outside world would fade completely as the only thing I would perceive would be the pain.  Wound up with back surgery to solve the problem. Didn't go away on it's own, physical therapy failed, Steroid injections into spine failed, so it was back surgery. 

They should do a nerve conduction test to help trace the problem.  If it's a pinched or damaged nerve they should be able to find out where it is exactly with that.

[Virmir]

Oh, yes, definitely one or two 10s in the beginning long before I started writing this down.  They both have reached an occasional seven or so since I started recording, but not for the past few months.  Essentially, I want to get back to no pain or nearly no pain at all.  If I knew once I started experiencing a three or four that it wasn't going to get any worse, I would have no problem barging ahead and working through the pain (I had no problem drawing comics and working on commissions through a couple of kidney stone and similar episodes).  However, I know from experience that pain in that range means I better stop what I'm doing or it's going to get much worse pretty quickly and stay that way for days. 

Plan... E or F (I'm losing count) is to just say forget it and draw anyway.  I'm confident I will be mostly fine this way, especially now since I can draw with both hands.  Although this is the very last resort.

I did have an EMG test to look for nerve problems and was surprised when they found nothing, especially given the problem with the right fingers.

Thanks for your comments!

[Fragmaster01]

When they did the EMG test, did they have you draw like you normally would? Or was it a lying down "here, now move this finger" sort of thing?

[Virmir]

I don't think they had me move my fingers at all.  It was just a bunch of electric shocks with my arm lying down that would cause my muscles to twitch.

[Fragmaster01]

That's no good. You need a test to replicate what actually happens to cause the pain. You mentioned that there's very little to no pain when not moving the limb, which makes me think that it's the motion causing it(nerves with shredded sheathing, pinched stuff, etc). I'd go have them hook you up and try and replicate the pain. Do the test so that they have you monitored while it's hurting.

I can ask a doctor on that for confirmation but yeah. Obviously they won't find any problems when it's not hurting. That wasn't very bright of them.

EDIT: I sent off an email to my aunt(hospital nurse/chaplain person) on the subject, to see if she knows or knows a neurologist at the hospital with advice.

[Virmir]

It's tricky because pain will usually come a little while afterwards, not when I initially do it.  I would have to draw or squeeze a ball or whatever for a little while to provoke it enough to start giving the warning signs, and then I would probably pay for it more later on in the day, probably especially that night.  (Although there have been some cases where I've gotten pretty quick responses.)

[Fragmaster01]

My research is limited, and will need more time/contact to work out, but this seems to be the most relevant:
https://en.wikipedia.org/wiki/Brachial_plexus_injury
It would be a minor form of it if it were that, but my main thoughts in that regard are on how you have similar responses in both arms. That suggests less a problem with the arms and more in the neck where they come together.

[Virmir]

That is interesting because I did have a strange episode of numbness in the right arm about one year before this happened although that just went away after a week or so.  This does suggest some sort of injury causing it though and nothing traumatic has happened, so I'm a bit doubtful there.  Still, good to keep in mind, though!

[Virmir]

This is the six month checkpoint I set for myself back at the end of December when I decided to cease all activity.  I am about halfway where I wanted to be at this point back then--that is just barely starting to get back into the comic (I completed one page this month!  Will hang on to this for a while before posting to see how future pages go and decide if I want to come back early at an extra slow pace, or just build the 10 page buffer first.) I wanted to be at this point by March or April or so, but because I am still seeing improvement, I will continue this schedule until I don't see improvement for a few months in a row or I actually get better.

This month's arm annoyance factors below! Again, keep in mind the right arm is nearly constantly at rest in the left arm in use. (Standard deviations in parentheses.)

Code: [Select]

L R
February 3.1 (1.4) 4.1 (1.8)
March 2.4 (0.9) 3.2 (1.3)
April 2.3 (0.9) 2.9 (1.0)
May 2.4 (0.9) 2.3 (0.9)
June 1.6 (0.6) 2.2 (0.8)
Right arm--perhaps slightly better compared to May, but not really enough to be significant.  That's okay because the previous improvement from April to May was fairly noticeable and it definitely did not get any worse.  I am going to keep the drawing schedule for the right arm the same this month (2 hours per week) as a precaution to see how it does.

Left arm--noticeably better!  It's funny because as soon as June started it started feeling better.  This is very good and unexpected since the past three months it has been about the same.  When it does get irritated, it seems to be more from heavy lifting or similar and not from drawing sessions or computer use so much lately (hence increased activity in chat).  I am going to increase the two weekday drawing sessions to 1 hour each and the Sunday session to *3 hours* to test the waters a bit, bringing the left handed drawing time up to 5 hours per week.  If these go well I might start doing left handed streams on Sundays.  If not, I will scale these back as this is a pretty big jump.

Extra Complex Fun Allotment Schedule:

Sunday - L 3h
Monday - R 30m
Tuesday - L 1h
Wednesday - (odd day)
Thursday - R 30m
Friday - L 1h
Saturday - R 1h

Total weekly drawing hours for July - 7

[Nixuelle]

Updated plots for July, considering the new data.  I only did the linear model for each arm this time, though I did add in confidence bands (90%) around the fitted line predictions.  As one might expect, the high standard deviations and the deviant data points result in a relatively poor extrapolation uncertainty on the fitted models.  As we get closer to approaching a pain factor of zero on each arm, the uncertainty should reduce considerably.